Corneal transplantation has evolved into a highly successful operation which can provide visual rehabilitation for a large variety of pathological conditions. Due to advances in tissue preservation and microsurgical technique, graft clarity can now be expected in up to 90% of uncomplicated transplant cases.1 High degrees of astigmatism, often coupled with high amounts of myopia, can cause significant visual disability in eyes with clear transplants. In relatively mild cases, spectacles may provide good visual correction for many eyes. Contact lenses may provide a more satisfactory optical correction in severe cases. They act to reduce both the aniseikonia from high myopia and meridional magnification resulting from high degrees of astigmatism. Unfortunately, contact lens wear can be limited by many factors. Patient intolerance or inability to handle and maintain contact lenses frequently occurs in elderly keratoplasty patients. High astigmatism or severe surface irregularity may preclude adequate contact lens fitting in these patients.
Corneal relaxing incisions, compression sutures, and wedge resections now provide the corneal surgeon with better means to treat postkeratoplasty eyes with high degrees of astigmatism when spectacles and contact lenses fail.212 These incisions induce a coupling phenomenon which flattens the steep axis along with steepening of the flatter meridian. As a result of this coupling effect, there is minimal change in the spherical equivalent refraction. Trapezoidal keratotomy, or the Ruiz procedure, offers a way to correct combined myopia and astigmatism. While this procedure has proven to be effective in reduction of astigmatism, it suffers from poor predictability and has a limited effect in reduction of myopia in postkeratoplasty eyes.13
Radial keratotomy has been shown to be a relatively safe and efficacious procedure used to reduce myopia in normal individuals.14 Recent work with four-incision radial keratotomy has shown reduction of myopia while potentially minimizing many side effects.1517 Four-incision radial keratotomy alone, or in combination with relaxing incisions in eyes with significant astigmatism, should provide a method to reduce myopia in keratoplasty patients who are unable to be visually rehabilitated by conventional means. This study provides a retrospective analysis of the efficacy of radial keratotomy in the reduction of myopia in corneal transplant eyes with high myopia. The interaction of radial keratotomy with relaxing incisions for astigmatism is also examined.
MATERIALS AND METHODS
We did a retrospective analysis of 11 keratoplasty patients who underwent four-incision radial keratotomy in a corneal graft for high myopia. Seven of these patients also underwent concomitant arcuate relaxing incisions in the graft-host interface for high astigmatism. In addition, one patient underwent paired tangential incisions in the graft for excessive astigmatism. All of these patients had ansiometropia or a high degree of astigmatism which resulted in either poor visual acuity with spectacle correction or the inability to wear a contact lens. The patients ranged from 34 to 82 years of age (mean, 66.1 years). Diagnosis before keratoplasty included: Pseudophakie corneal edema, keratoconus, Fuchs* corneal dystrophy, and corneal scarring due to Herpes simplex virus. The average time from penetrating keratoplasty to radial keratotomy was 4.1 years (range, 6 months to 9 years). The postoperative follow up ranged from 3 months to 5 years with an average of 16 months.
The surgical procedures were all performed under an operating microscope using standard operative sterile technique. Topical anesthesia was achieved by the use of proparacaine 0.5% drops given three times, 5 minutes apart. Topical antibiotics were also given prior to starting the procedure. The central cornea was marked with a Sinskey hook while the patient maintained fixation on the filament of the operating microscope. An optical zone marker (3 or 4 mm) was then used to mark the central cornea. Ultrasonic pachometry was then used to measure the central and paracentral corneal thickness in each of the four quadrants. A diamond knife was set to 100% of the thinnest paracentral reading. The globe was then fixated and a front cutting diamond knife was used to create four radial incisions starting just anterior to the graft-host interface extending to the central optical zone. In one case (#8), a back cutting knife was used to make incisions extending from the central optical zone to graft-host interface.
When relaxing arcuate incisions were used in conjunction with the radial keratotomy technique, the radial keratotomy incisions were oriented 45° from the major corneal meridians as determined by keratometry. This allowed the radial keratotomy incisions to straddle the arcuate incisions which were centered at the steepest axis.
After completion of the radial keratotomy incisions, arcuate intraincisional relaxing incisions were used for eyes with high degrees of residual astigmatism. For these incisions, a #75 blade or a diamond knife was used to create a three-quarters depth incision in the wound at the graft-host interface. The incisions were paired and were centered around the steep axis extending from 45° to 90° depending upon the degree of astigmatism. One patient (#4) received a single, 90-degree unpaired arcuate incision. Another patient (#7) received paired, 3-millimeter straight tangential incisions in the graft at a 6-millimeter optical zone, cut with a diamond knife. At the completion of the procedure, the incisions were irrigated and antibiotic drops applied.
Shown in Table 1 are the patient age, preoperative diagnosis, preoperative refraction and best spectacle corrected visual acuity, surgical procedure performed, postoperative refraction, and corrected visual acuity and length of follow up. All refractions are listed in positive cylinder form. Nine patients showed improved or stable visual acuities following the procedure. Two patients showed worsening of corrected visual acuity postoperatively. This was felt to be due to: increased astigmatism, with induction of some irregular astigmatism following relaxing incisions in one patient (#5), and worsening of preexisting cystoid macular edema in another patient (#3). Table 2 shows the relationship between preoperative and postoperative spherical equivalent refraction, keratometric astigmatism, and refractive astigmatism. All 11 patients demonstrated a reduction in myopia from the preoperative state. The mean preoperative spherical equivalent was - 6.46 diopters (range, - 1.50 to - 13.00 D) and the mean postoperative spherical equivalent was -2.98 D (range, - 1.00 to - 5.38 D). The average reduction in spherical equivalent refraction was 3.48 D (range, 0.25 to 7.75 D). Seven out of eight patients who underwent relaxing incisions or transverse incisions for high astigmatism demonstrated a reduction in postoperative keratometric astigmatism. The average preoperative keratometric astigmatism for this group was 5.72 D (range, 3.50 to 8.50 D) with the average postoperative value of 3.47 D (range, 1.00 to 10.00 D). The mean reduction of keratometric astigmatism for these patients was 2.25 D (range, - 6.75 to +6.50 D). The mean preoperative refractive astigmatism for the group of eight patients undergoing astigmatic surgery was 6.94 D (range, 3.5 to 10.5 D) with a mean postoperative value of 2.31 D (range, 0.25 to 6.0 D). The average decrease in cylindrical refraction was 4.63 D (range, 1.5 to 9.5 D). The keratometric values for the three patients undergoing radial keratotomy alone are also listed. The mean preoperative keratometric astigmatism for this group was 1.83 D and the mean postoperative value was 3.00 D with an average difference of + 1.17 D. One patient who underwent radial keratotomy alone showed an increase of 1.00 D of astigmatism in his postoperative correction. Overall, the group of three radial-keratotomy-only patients had a reduction in refractive astigmatism of 0.33 D.
Results of a Refractive Keratotomy After Penetrating Keratoplasty
Two patients had microperforations, one occurring in a relaxing incision (#6) and one in a radial keratotomy incision (#9). Both resolved spontaneously, without requiring suturing. One patient (#3) developed a small epithelial defect on the donor cornea adjacent to a radial keratotomy incision near the graft-host junction. This resolved with treatment with lubricants and a small temporary lateral tarsorrhaphy. It should be noted that there was no subjective postoperative glare disability noted by any patient.
Postoperative refractive errors occur frequently in keratoplasty eyes and can limit visual rehabilitation in patients with clear transplants. Unpredictable changes in corneal topography after surgery, often coupled with difficulty in intraocular lens power calculation in triple procedures, can contribute to development of these refractive errors. In one series, up to 38% of keratoplasty patients undergoing triple procedures had refractive errors greater than 2.00 D from emmetropia.18 The majority of these patients can achieve good visual rehabilitation by conventional means. Spectacle correction involving moderate amounts of anisometropia and relatively high degrees of astigmatism are often surprisingly well tolerated in transplant patients.19 Contact lens correction minimizes aniseikonia and irregular astigmatism and can be used in more severe cases. Patient intolerance often limits contact lens use in elderly keratoplasty patients.
Change in Refraction and Keratometry Induced by Refractive Keratotomy After Penetrating Keratoplasty
Radial keratotomy offers a relatively safe and efficacious way to reduce myopia in normal individuals.14 This procedure has undergone many changes and refinements since its introduction in the United States in the late 1970s. The current trend is toward the use of fewer incisions than had been initially popular.1517 Four-incision radial keratotomy offers many potential advantages over procedures involving more incisions. These include: increased corneal stability, decreased risk of perforation, less chance for overcorrection, less glare, less visual fluctuation, and lower potential for endothelial cell loss.15 The disadvantage of this procedure is that there may be a trend toward more frequent initial undercorrection. Undercorrections can easily be treated by extending, redeepening, or adding corneal incisions at a later time when the wounds have stabilized. The predictability of radial keratotomy has also improved over time. Refinements in surgical technique have improved technical precision while intensive statistical analysis has provided a variety of regression formulas and nomograms to aid in preoperative planning. A variety of factors are felt to influence the outcome of radial keratotomy surgery in normal individuals. The number and depth of incisions, size of the optical zone, and patient age are the primary factors.20 In addition, patient sex, keratometry readings, corneal thickness, and intraocular pressure can also influence the final results.20 In our retrospective analysis, four-incision radial keratotomy reduced the spherical equivalent myopic correction by an average of 3.48 D. There was a high degree of variability, ranging from 0.25 to 7.75 D. There were no overcorrections. Nine cases utilized a 3millimeter optical zone, while a 4-millimeter optical zone was used in two cases. The choice of optical zone size was determined on an individual case basis by surgeon experience since no previous radial keratotomy data was available from keratoplasty patients. The reduction of myopia was greater in eyes with a smaller optical zone (3.74 D with 3-millimeter optical zone, 2.32 D with 4-millimeter optical zone). Additional variables in the keratoplasty eye which may complicate or influence predictability include: time postkeratoplasty, graft size, recipient site size, donor age, graft sex, initial indication for keratoplasty (especially in cases of keratoconus or ectatic dystrophies), surface topography, reduced incision length (due to peripheral graft border).
It should be emphasized that the radial incisions did not extend peripherally beyond the graft-host interface. This technique allows preservation of the recipient bed for future keratoplasty, if necessary, and avoids potential problems with wound healing or gape due to intersection of the incisions with the wound interface. We would expect that the shortened corneal incisions would induce less corneal flattening than incisions extending closer to the limbus. It is difficult to judge the magnitude of this effect since the peripheral segment of a radial incision is felt to contribute less to corneal flattening than the central segment.20 It is also notable that in the three patients that underwent isolated radial keratotomy there was a mild increase in keratometric astigmatism (1.17 D) while refractive astigmatism was slightly reduced ( - 0.33 D). This probably reflects the dynamic nature of the transplanted cornea and could potentially effect planning for astigmatic correction.
Attention must also be directed toward potential complications of radial keratotomy in the keratoplasty patient. These include induction of irregular astigmatism, vascularization, weakening the structural integrity of the graft, fluctuating visual acuity, glare, keratitis, and epithelial erosions.21 One patient in our series manifested a small, transient epithelial defect which resolved with lubricants and a temporary tarsorrhaphy. Two patients had selfsealing microperforations during the procedure which required no intervention. One additional patient had reduction of myopia and astigmatism, but decreased best spectacle corrected visual acuity due to worsening of preexisting cystoid macular edema. Best spectacle corrected visual acuity was also reduced in one patient due to the development of irregular astigmatism despite reduction of myopia and keratometric astigmatism. Overall, there was a reduction of best spectacle corrected visual acuity in 2 of the 11 patients despite reduction of myopia and keratometric astigmatism.
Arcuate relaxing incisions have been shown by several groups to be an effective way to reduce astigmatism in postkeratoplasty eyes.2"12 Previous studies have shown that paired arcuate intraincisional relaxing incisions show a coupling effect causing both a flattening of the steep axis and a steepening of the flat axis.2 These have been shown to occur in an equal ratio and, therefore, should not affect the spherical equivalent or the amount of predicted flattening.2 A similar coupling effect has been shown to occur with paired straight tangential incisions.22 In our study, 7 of 11 patients had high amounts of astigmatism (> 3.50 D) and underwent both radial keratotomy and arcuate relaxing incisions. One additional patient underwent radial keratotomy with concomitant paired transverse incisions. These patients showed a mean reduction of keratometric astigmatism of 2.25 D and refractive astigmatism of 4.63 D. This is in comparison to our previous work which showed a mean reduction of 4.70 D of keratometric astigmatism using 90-degree paired arcuate relaxing incisions.2 In the current study, arcuate incisions varied from 45° to 90°, depending upon the degree of astigmatism. All eyes showed a reduction of keratometric astigmatism in the steepest preoperative axis. There was considerable variation in individual effect with a range of - 6.75 to + 6.50 D. This variability is consistent with that noted in previous studies.2·22 One patient (#5) had a net increase in keratometric astigmatism due to production of excessive steepening in the previously flat axis. It should be noted that there was some discrepancy between the values measured by refractive and keratometric astigmatism. This has previously been noted in keratoplasty patients by other investigators.23 Complicated postoperative corneal topography and limitations of keratometric analysis, which measures only two isolated points near the central 3millimeter optical zone, could possibly account for some of these differences. In addition, highly myopic eyes frequently manifest greater amounts of keratometric astigmatism than astigmatism measured at the spectacle plane. We feel that the refractive astigmatism represents a practical way to evaluate the functional status of the graft. Future analysis, using computerized topographical systems may help to elucidate these differences.
In summary, it appears that radial keratotomy offers a tool for the corneal surgeon to treat keratoplasty eyes with high degrees of myopia and resulting anisometropia that cannot be visually rehabilitated by conventional means. The modified four-incision technique showed reduction of myopia in all eyes. Shortened radial incisions avoid crossing the graft-host interface and minimize potential wound healing difficulties and preserve the donor bed. It appears that radial keratotomy can be combined with arcuate relaxing incisions to reduce astigmatism in complicated cases. Tb our knowledge, this is the first reported series describing the effect of radial keratotomy in keratoplasty eyes. At this point, predictability suffers from small patient numbers which limit statistical analysis and the added variability and instability of the corneal surface in the keratoplasty eye. Development of irregular corneal astigmatism, with reduction of visual acuity is a potential side effect in these eyes. The goal in these cases should be to reduce myopia to lower levels which are acceptable for better visual function with conventional corrective measures. Future advances in excimer laser technology may provide a more precise and predictable way to eliminate myopia in keratoplasty eyes.
1. Bourne WM. Current techniques for improved visual results after penetrating keratoplasty. Ophthalmic Surg. 1981;12:321.
2. Lavery GW, Lindstrom RL, Hofer LA1 Doughman DJ. The surgical management of corneal astigmatism after penetrating keratoplasty. Ophthalmic Surg. 1985;16:165-169.
3. Belmont SC, Troutman RC Compensating compression sutures in wedge resections. Journal of Refractive Surgery. 1985; 1:104.
4. Troutman RC, Swinger C. Relaxing incision for control of postoperative astigmatism following keratoplasty. Ophthalmic Surg. 1980;11:117.
5. Sugar J, Kirk AK. Relaxing keratotomy for postkeratoplasty high astigmatism. Ophthalmic Surg. 1983;14:156.
6. Krachmer JH, Ching SST. Relaxing corneal incisions for postkeratoplasty astigmatism. Int Ophthalmol Clin. 1983;23:153.
7. McCartney DL, Whitney CE, Stark WJ, et al. Refractive keratoplasty for disabling astigmatism after penetrating keratoplasty. Arch Ophthalmol. 1987;105:954.
8. Mandel MR, Shapiro MB, Krachmer JH. Relaxing incisions with augmentation sutures for the correction of postkeratoplasty astigmatism. Am J Ophthalmol. 1987;103:441.
9. Kelley CG, Kaufman HE. Relaxing incisions postkeratoplasty: a new technique. Invest Ophthalmol Vis Sci. 1984;25:27.
10. Limberg MB, Dingeldein SA, Green MT, Klyce SD, Insler MS, Kaufman HE. Corneal compression sutures for the reduction of astigmatism after penetrating keratoplasty. Am J Ophthalmol. 1989;108:36-42.
11. Agapitoe PJ, Lindstrom RL, Williams PA, Sanders DR. Analysis of astigmatic keratotomy. J Cataract Refract Surg. 1989;15:13-18.
12. Lindstrom RL, Lindquist TD. Surgical correction of postoperative astigmatism. Cornea. 1988;7:139-148.
13. Merck MP, Williams PA, Lindstrom RL. Trapezoidal keratotomy: a vector analysis. Ophthalmology. 1986;93:719-726.
14. Waring GO, Lynn MJ, Fielding B, et al. Results of the Prospective Evaluation of Radial Keratotomy (PERK) Study 4 years after surgery for myopia. JAMA. 1990;263:1083-1091.
15. Spigelman AV, Williams PA, Nichols BD, Lindstrom RL. Four-incision radial keratotomy. J Cataract Refract Surg. 1988;14:125-128.
16. SaIz JJ, Villasenor A, Elander R, Reader AL, Swinger C, Buchbinder M. Four-incision radial keratotomy for low to moderate myopia. Ophthalmology. 1986;93:727-739.
17. Spigelman AV, Williams PA, Lindstrom RL. Further studies of four-incision radial keratotomy. Refract Corneal Surg. 1989;5:292-295.
18. Crawford JC, Stulting, RD, Waring GO, Van Meter WS, Wilson LA. The triple procedure: analysis of outcome, refraction and intraocular lens power calculation. Ophthalmology. 1986:93:817-824.
19. Mannis MJ, Zadnik K. Refracting the corneal graft, Surv Ophthalmol. 1990:34:436-440.
20. Hofmann RF, Lindstrom RL. Standard surgical technique for single pass four and eight incision radial keratotomy. In: Sanders DR, ed. Radial Keratotomy: Surgical Techniques. Thorofare, NJ: SLACK Ine; 1986:35-64.
21. Rashid ER, Waring GO. Complications of radial and transverse keratotomy. Surv Ophthalmol. 1989;34:73-106.
22. Lindstrom RL. The surgical correction of astigmatism: a clinician's perspective. Refract Corneal Surg. 1990;6:441-454.
23. Judge D, Gordon L, Vander Zwaag R, Wood TO. Refractive versus keratometric astigmatism postkeratoplasty. Refract Corneal Surg. 1990;6:174-178.
Results of a Refractive Keratotomy After Penetrating Keratoplasty
Change in Refraction and Keratometry Induced by Refractive Keratotomy After Penetrating Keratoplasty